One Published Decision — Can you be actually innocent of a crime you weren’t convicted of?

Editorial Comment: Today we have a classic example of how the American Criminal Justice System can ensnare people — both the guilty and innocent — in a web of contrived evidence, pressure to convict, and grossly ineffective assistance of counsel — and then fails to provide any kind of real relief to those whose lives were destroyed in the process — or closure for the victim’s family and friends. I urge you to read the entire post, because it’s important that everyone, not just attorneys, understand that the system, while often just, is not perfect.

The Court of Appeals issues a single published decision today, an order dismissing a petition for a writ of actual innocence.  The issue addressed by the Court is not whether Stephen James Hood was actually innocent of being an accessory after the fact to abduction and first-degree murder of Ilouise Cooper, and elderly black woman, as a principal in the second degree, but whether the Court can consider his petition where Hood was technically not convicted of these offenses.

The murder occurred August 31, 1990. Initially, police believed that Cooper had been murdered by Billy Madison as revenge against a relative or friend of Cooper who had beaten and robbed Madison during a drug deal gone wrong.  Hood, an acquaintance of Madison, was thought to have aided in the crime. Eventually, police brought in witnesses to identify Madison and Hood from a photo line-up.  Also in the line-up was Jeffrey Cox, also an acquaintance of Hood and Madison.  The witnesses gave very tentative identifications of Cox as the person who abducted Cooper.

Believing that Cox was the principal actor — but also that Hood was an accomplice — prosecutors offered Hood immunity in exchange for his testimony.  Hood was presented with “statements” that he and Cox had abducted and murdered Cooper, and at the urging of his attorney, he adopted these statements as his own.  The theory of the crime was somewhat vague, with the Commonwealth seemingly unsure whether the abduction was with the intent to rob Cooper or sexually assault her, or both.  This theory was complicated by the fact that there was neither evidence or robbery — her jewelry was not taken — nor of sexual assault — although her close were “in disarray,” there was no evidence of trauma suggestive of a physical violation.

Cox was convicted of abduction and murder, and his appeal to the Court of Appeals was refused.  Cox’s attorney failed to seek an appeal of that decision to the Supreme Court of Virginia.  This was back in the day when a delayed appeal could only be granted by filing a habeas corpus which had to include all other bases for overturning the conviction or they would be deemed waived.  Cox, perhaps wisely as it turned out, elected not to seek a delayed appeal at that time.  He then spent the next decade in an effort, both pro se and with the assistance of counsel, seeking to challenge his convictions. I will not go into the details of these efforts that led to Cox’s eventual exoneration, which you can read about in the National Registry of Exonerations. Let it suffice to say that the evidence against Cox was shown to be a tissue of lies and that Cox had an airtight alibi for the time of the abduction and murder.  After his release from prison, the General Assembly passed a compensation bill awarding him $750,000.

And that should have ended this sorry affair.  Except, the Commonwealth still had Hood’s “confession.”  Theorizing that Hood had agreed to the immunity deal to cover up his guilt as the actual murderer, the prosecution indicted Hood and successfully convinced the circuit court to throw out the immunity deal on the basis that Hood’s “statements” implicating Cox were perjured. The Commonwealth also persuaded the court to prohibit Hood from presenting any evidence with respect to Cox’s conviction and exoneration to a jury.  As a result, Hood opted for a bench trial,

Now the case gets even murkier, as the Commonwealth’s theory was that Hood was the principal actor, but the evidence actually implicated Madison, with hood as an accomplice, because the Commonwealth had reverted to “revenge for a drug deal gone wrong” theory, and Madison was the victim of the robbery and beating.  The circuit court rejected Hood’s claim that his initial “confession” implicating Hood had been coerced, but struck the more serious charges and convicted him as a second degree principal in the murder and an accessory after the fact to the abduction. Hood filed an appeal to the Court of Appeals, which upheld his conviction in a 2-1 decision with Judge James Benton dissenting. The Supreme Court of Virginia affirmed the decision of the Court of Appeals.

We pause now in our main story, the bring in some relevant facts.  Anyone who was living in Richmond in the mid-90s probably remembers the investigation into the Golden Years Murders.  For the better part of a decade, elderly black women were abducted, sexually assaulted and murdered, the work of an apparent serial killer.  A special task force was formed to investigate the crimes.  Eventually, Leslie Leon Burchart, a mentally ill homeless man, was convicted of several of those murders.  Why is this relevant?  Because it was subsequently alleged, with very credible support, that Burchart was probably guilty of additional murders — including Coopers.  While this was far from certain (and there was much stronger evidence in another case that a person had been wrongfully convicted of a murder likely committed by Burchart), it was clear that Richmond Police had not pursued leads in these cases that would have linked Burchart to the murders.

After Burchat was convicted, Hood was sought a writ of habeas corpus from the Circuit Court of the City of Richmond.  It took the court nearly five years to resolve the petition, eventually concluding that Hood’s trial had been a clusterf**k of errors by his counsel and likely misdeeds by the prosecution.  Predictably, the Commonwealth sought to appeal this decision.  However, the appeal was subsequently withdraw when Hood agreed to enter an Alford plea to attempted abduction with a sentence of time-served.  It is easy to understand why Hood would do this.  Even if the Commonwealth’s appeal of the habeas had been unsuccessful, Hood was likely facing a retrial in which, given his past experience, his confidence of an acquittal was understandable low.

Which brings us to his petition for a writ of actual innocence.  Hood was asking the Court of Appeals to find that he as actually innocent of the convictions that were vacated by the habeas order. While this would not have resulted in the vacating of his conviction under the Alford plea, it would have gone a long way to acknowledging that he was likely the victim of an overzealous prosecution.

The problem is, as Judges Humphrey and Causey and Senior Judge Clements see it, is that the Court does not have jurisdiction to consider a writ of actual innocence under these circumstances.  Recognizing that this is unique situation, the Court analyzes its statutory jurisdiction and concludes that writs of actual innocence are available only to set aside a conviction of a felony offense — the accessory after the fact offense was, at that time, a misdemeanor and so wouldn’t be reviewable in any case, as pointed out in a footnote — but the convictions have been vacated, rendering them a legal nullity.  As the convictions do not legally exist, the Court concludes that it has no jurisdiction to determine is Hood was actually innocent of the crimes.

I have not fault with the reasoning of the Court — the statute permitting writs of actual innocence just didn’t anticipate this type of case, not is it likely the General Assembly would have given such authority to the Court of Appeals had it been anticipated. My issue with the case is more fundamental.  Too often I have seen criminal cases, a prime example being the repeated attempts to impose the death penalty on Darryl Renard Atkins, where the Commonwealth pursued a defendant relentlessly despite increasing evidence of his likely innocence.

In the Atkins case, it was ultimately revealed that Atkins, while an accomplice to the robbery of a young sailor and present at his subsequent murder, had been railroaded by an unscrupulous prosecutor who determined that because Atkins was developmentally disabled a jury would be more likey to impose a death sentence on him than on his accomplice.  The prosecutor then coached the accomplice to change his statements to implicate Atkins.

I will not recount the long and tortured process that followed, but will say this: I was responsible for reviewing the evidence from Atkins’ first trial — before the prosecutorial misconduct was revealed.  After studying the physical evidence and the testimony of the accomplice, I was convinced beyond any doubt that the accomplice had been the shooter and, in all likelihood, Atkins had tried to prevent the murder. Unfortunately, by this time, Atkins “guilt” had been established and the only issue was whether the Commonwealth could put to death a developmentally disabled person without giving the jury an opportunity to weight this as a mitigating factor.  The case went all the way the the United States Supreme Court, which reversed the death sentence.  The Commonwealth was able to obtain a death sentence from a second jury, but that sentence was also set aside, this time by the Supreme Court of Virginia.  It was while Atkins was awaiting his third penalty proceeding that his accomplice dies in prison, and the accomplice’s attorney, freed from the attorney-client privilege, revealed that his client had admitted his role as the shooter and that the Commonwealth, aware of this claim, had nonetheless proceeded to use his testimony against Atkins.

With this revelation, the trial judge ruled that no reasonable jury would have convicted Atkins as a principal in the first degree, which is required to impose the death sentence, and entered a judgement sentencing Atkins to life in prison.  The Commonwealth then did something that I find to be inexcusable — it sought a writ of mandamus to the trial court directing it to vacate the sentence and to permit the Commonwealth to seek a third death sentence from a jury.  The Supreme Court of Virginia refused the petition.

The present case is perhaps not as grave as the Atkins case, it is still incomprehensible to me that that Commonwealth would continue to seek a conviction against Hood, who spent nearly a decade in prison, given the absolute certainty that obtaining a conviction against him was a dicey prospect at best.  Unlike in Atkins, I have not had the opportunity to review the evidence first hand and in detail, and I suppose it is possible that there was some credible evidence that linked Hood to the abduction and murder of Cooper.  But I do not see the conviction of Hood by an Alford plea to “attempted abduction,” a crime that he most certainly did not commit, as Cooper was actually abducted, as in any way constituting “justice.”

Failure to Make a Sufficient Proffer has Sunk Many Appeals . . . But What Makes an Proffer Sufficien

When testimonial evidence is excluded by a trial court, it is essential that the party who want to introduce that evidence makes a “proffer” of what the evidence would have been.  Without a proffer, an appellate court will rarely consider reserving a judgment based on the trial court’s decision to exclude the evidence as there is no way of knowing how it would have impacted the trial.  Perhaps the witness would have claimed ignorance, or perhaps would have said the exact opposite of what the appellant wanted.

Now when a court excludes documentary evidence, making a proffer is easy.  You simply produce the documents and have them entered into the record.  Generally, the court will not even require you to authenticate them since they are not being considered.  But testimonial evidence is not that easy to proffer for several reason.  First, the court isn’t likely to allow you to proffer it through live testimony.  Whether it is a bench trial or a jury trial, this will take time, and in the case of the former, there is the added difficulty that the judge, who is also the trier of fact, will have to be present when the testimony is proffered, but ignore it later when making determinations of fact.  Judges often say, “You can make your proffer after the evidence is in,” but this requires 1) that you remember to do it, and 2) that the witness is going to be available at the end of the trial.  There are also times when you are being asked to proffer testimony of a witness who is not present, or to proffer the expected testimony rather than have the witness testify.

The making of a proffer must be sufficient to allow the appellate court to determine if its exclusion was error, and that is the issue that controls the result in today’s only published opinion from the Court of Appeals, William O. Flannagan, Jr. v. Commonwealth of Virginia. Flannagan was charged with and convicted of 1st degree murder and use of a firearm in that crime. It was not disputed that on the day of the murder Flannagan had been drinking at a gathering and over the course of several hours got into repeated arguments with Jason Ferguson, at times recklessly brandishing a pistol.  Eventually, Flannagan and Ferguson were both asked to leave.  Shortly thereafter a gunshot was heard and Ferguson was shot and killed.  Although Flannagan was standing nearby with the gun in his hand, apparently no one actually saw Flannagan fire the pistol.

Flannagan walked away from the scene and was apprehended a short time later.  The officers who found Flannagan administered a preliminary breath test (PBR).  A PBT is generally used to detect the presence of alcohol in a person’s breath, but is not recognized as scientifically accurate to provide the precise concentration of alcohol.  In most cases, a PBT is used to establish probable cause that a person is intoxicated so that a more accurate test can be be ordered, but by statute, the result of the test is not admissible to establish the blood alcohol level (BAL) of a defendant during the guilt phase of a trial.  This rule normally benefits the defendant in a DUI case or for some other offense where a precise BAL is required to be shown for some purpose, but it applies to both the prosecution and defense — so, for example, the defense cannot seek to introduce the result of a PBT to rebut a BAL established by scientifically accurate means.

The use of PBR results in other contexts is less certain.  Certainly, if the result is to be introduced to show that the person to whom the test was administered had some alcohol in his blood stream, for example to show that he had been surreptitiously been given an alcoholic beverage,  it could be argued that the test was sufficiently accurate to prove this.  Attempts have been made to have PBR tests admitted in both criminal a civil cases where the issues was not the test’s ability to show a precise BAL, but only to show that some alcohol was present.  In such cases, the courts have generally based the admissibility on two factors: 1) Is the particular PBR test accurate when properly calibrated and administers and 2) that it was properly calibrated and administered in the particular case.

Now you are probably wondering (unless you know a lot about homicide law) why the result of Flannagan’s PBT would have been relevant in this case.  He was charged with first degree murder, not DUI.  The answer lies in a peculiar facet of criminal law in Virginia, which is that voluntary intoxication is not a defense to negate or mitigate the existence of an intent to commit a crime except to negate premeditation as an element of first degree murder.

Flannagan wanted to question the officers who administered the test about the specific result of the test, that is, the approximate BAL that the test indicated was in his system just shortly after the killing.  While the evidence would establish that Flannagan had been drinking, that was probably not enough to show that he was intoxicated to the point that his ability to premeditate the killing of Ferguson would be called into doubt.  Flannagan obviously wanted to put some number to his level of intoxication to bolster his defense, and the Commonwealth just as obviously wanted that evidence out.

When the Commonwealth objected to Flannagan’s first question to the officer about the PBT, Flannagan said that he would present expert testimony regarding what the result meant with respect to his level of intoxication.  His counsel proffered that the officer who administered the test would say that it had been calibrated just five days before and would also authenticate the calibration logs for the test.  The circuit court then asked how that “establish[ed] the reliability of the test overall.” Citing Santen v. Tuthill, 265 Va. 492 (2003) as well as an Attorney General opinion, Flannagan maintained that the proper calibration of a PBT device was sufficient to establish its reliability.

The court seemed doubtful of this proposition and asked if the expert would state that she could render her opinion based on the device being properly calibrated because the test was “reliable for establishing the blood alcohol content.”  Following a brief recess, Flanngan’s counsel reported that the expert was not prepared to say that the device was “reliable for establishing the blood alcohol content.” The court ruled that the evidence of the result of the test, and the testimony of the expert based on that result, were not admissible, but permitted Flannagan to proffer the expected testimony, which was that the the device was properly calibrated and administered, that the PBT showed a BAL of 0.189%, and that this level of intoxication would have affected Flannagan’s “judgment, attention, motor coordination, and reaction time.”

The issue on appeal was whether the proffered evidence should have been admitted.  More specifically, Flannagan argued that the proffer that the device was properly calibrated and the test properly administered was all that was required to establish the reliability of the result.  In an opinion by Judge AtLee, joined by Judge Causey and Sr. Judge Haley, the Court of Appeals disagrees.  Setting aside for a moment that Santen is a civil case, so probably not the best precedent for admissibility of this kind of evidence in a criminal trial, the Court points out that the circumstance in this case is the mirror image of that in Santen. In Santen the proffer was that the device was scientifically accurate, but no proffer was made that it had been properly calibrated and administered.  Here, the proffer was that the device was calibrated and the test administered properly, but not that the test itself was scientifically reliable.  As mentioned above, the general rule before admitting such evidence is that you must show both prongs — and Flannagan’s proffer didn’t.

Now, one could wonder (indeed, I’d be surprised if you aren’t wondering now), how can one expert testify in one trial that a test is scientifically accurate and another then not be able to do so in a different — for clearly that was the problem here.  There are several possibilities.  First, it could be that Flanngan’s expert simply wasn’t prepared to say this — that doesn’t mean that she could not have done so had she studied the literature or performed tests on the machine, just that she wasn’t able to say so on the day of trial.  Another possibility is that this could be an issue on which there is genuine scientific disagreement, because where some process is so widely accepted in the scientific community that there is no debate about its accuracy anymore, the courts will accept that fact without expert support.  In the case of the various types of PBTs, however, there is no consensus as to their precision of reliability — that is, while it is agreed that they can detect the presence of alcohol in the breath, unlike more sophisticated devices, the ability to accurately determine BAL is far from certain.

An Appellate Lawyer Goes to Trial . . . That Sounds Mental (and it is)

Today I second-chaired a criminal trial.  Why was an appellate lawyer doing real lawyer work?  Well, primarily because in this case (and in a related case that will be tried in about 2 months) there was initially a concern that there would be a significant appellate issue.  As it turned out today, we never even reached that issue and got an appropriate result. For reasons of confidentiality (although this was a public trial, I do not wish to bring any more attention to the case than is warranted), I won’t go into to details.  However, the issue we thought might need to be preserved for appeal was whether Virginia’s new “mental health evidence” law has opened the door, if just a crack, for obtaining an acquittal or a reduced offense based on the defendant’s diminished capacity/culpability.

For those not in the know, until July 1, 2021, in Virginia you could not introduce evidence of the defendant having a mental disease, mental defect, or developmental delay during the guilt determination of the trial unless you were seeking an acquittal by reason of insanity.  In Virginia, an insanity defense is applicable only if at the time of the offense(s) the defendant was “incapable of understanding the wrongfulness of his act,” which is often called the M’naughten Rule, or the defendant was acting under an “irresistible impulse.”

The M’naughten rule was formulated as a reaction to the acquittal in 1843 of Daniel M’naghten on the charge of murdering Edwaed Drummond. M’naghten had shot Drummond after mistakenly identifying him as UK Prime Minister Robert Peel, who was the intended target. The House of Lords asked a panel of judges, presided over by Sir Nicolas Conyngham Tindal, Chief Justice of the Common Pleas, a series of hypothetical questions about the defense of insanity. The principles expounded by this panel have come to be known as the “M’Naghten Rules”, though they have gained any status only by usage in the common law and M’Naghten himself would have been found guilty if they had been applied at his trial.

Although the panel offered several rules, the one that has received the most widespread use relates specifically to whether a defendant can be held liable for an action that he did not perceive to be wrong.  The judges said that “the jurors ought to be told in all cases that every man is to be presumed to be sane, and to possess a sufficient degree of reason to be responsible for his crimes, until the contrary be proved to their satisfaction; and that to establish a defense on the ground of insanity, it must be clearly proved that, at the time of the committing of the act, the party accused was laboring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing; or, if he did know it, that he did not know he was doing what was wrong.” For much the remainder of the 19th century and through the first half of the 20th century, this standard for insanity became the rule in almost all courts that derive their procedures from English common law, including most states in the US.

A second test, somewhat more recent in origin, is the “irresistible impulse test” in which the defendant, while aware that what he is doing is criminal, is compelled to commit the offense.  This test is sometimes called the “cop at the elbow” scenario, essentially maintaining that the defendant would have committed the crime even if the police were standing right beside him telling him it was a crime.

Apart from these two circumstances, mental illness was not a basis for exculpating a defendant of criminal liability, but could only be used to mitigate his punishment.  Indeed, in four US states and in many other criminal justice systems, the concept of using insanity as a defense is rejected flat out, allowing only for consideration of the punishment.

By the latter half of the 20th century, criminal justice reform had started to allow for consideration of mental illness to mitigate, or even exculpate, a defendant even when the defendant understood the wrongfulness of the act and was not action under an irresistible impulse.  Though known by various names, this practice has generally been referred to as the “diminished capacity” defense.  In very simple terms, its presents the argument that the defendant’s mental illness kept him from forming a rational intent to commit the crime, even while knowing that it was a crime.

Virginia has never adopted a diminished capacity defense, but this did not stop defense attorneys from trying to bring in evidence of a defendant’s mental illness even where an insanity defense was not being offered. So in 1985, the Supreme Court of Virginia was asked to decide whether such evidence was admissible in the guilt phase of a criminal trial. In Stamper v. Commonwealth, the Court held that such evidence was not relevant to proving the guilt or innocence in absence of a claim of insanity.  As the Court put it, “there is no sliding scale of insanity.”

And so the law remained until the 2021 Special Session of the General Assembly.  Taking up a bill that had been continued from the regular session, the legislature “overruled” Stamper by creating a specific statutory authorization under Code § 19.2-271.6 to admit evidence of a mental disease or defect or a developmental disability if such evidence tended to show that the defendant “did not have the intent required for offense charged.”  The evidence has to be otherwise admissible, and it must be established that the defendant has the disease, defect or delay, although the manner for doing so is not clearly set out in the law. The law also does not clearly require a court to conduct a commitment proceeding if the defendant is acquitted, though it makes reference to the court being permitted (or rather not being limited in its authority) to enter an emergency custody order and schedule such proceeding.

So what is the significance of this new law?  No one really knows, and that is why an appellate lawyer was second chairing a felony trial today, because there was a real possibility that the issue was going to come up.  As it turned out, it didn’t, because the trial court found that the evidence was sufficient to find that defendant was legally insane at the time of the offenses under the M’naughten rule.  While the court expressed some concern about one of the charges, it ultimately ruled that while the defendant had the burden of establishing insanity, the Commonwealth still retained the burden of proving guilt beyond a reasonable doubt, and in this instance the evidence that the defendant was mentally ill was sufficient to raise a doubt as to whether, absent that illness, he would have acted as he did, the court concluded that in his mental state it was not shown that he did understand the wrongfulness of his act, and as one witness opined that he did not, the defense prevailed.

So for the moment, we will not have a case on the application of Code § 19.2-271.6 to take up on appeal.  But there will have to be one, eventually.  The best possible interpretation (from the defense perspective) would be to treat the evidence as admissible to establish a “but for” principle. That is, “but for the mental illness, the defendant would not have committed the crime.”

For example, suppose the defendant “John Doe” believes himself to be Willie Sutton, the bank robber who when asked why he robbed banks, famously replied “Because that’s where the money is.” (Sutton denied having said it, saying that the reporter who attributed it to him made it up, but later said that had he been asked the question he might have said that, but would more likely have replied, “Because I enjoyed it.”).  Now Sutton knew that robbing banks was a criminal offense, and so does the defendant.  He commits the crimes because he enjoys it, or perhaps because that’s where the money is.  But he is not Willie Sutton, and if he wasn’t mentally ill, he wouldn’t ever dream of robbing a bank.

Under the M’naughten rule, an expert would have to opine that John Doe didn’t appreciate the wrongfulness of his acts when he believed that he was Willie Sutton and it was Sutton, not Doe, who was robbing banks.  The defense might find an expert willing to make this very fine distinction — i.e. that while John Doe as Willie Sutton knew robbing banks was a crime, he wasn’t able to appreciate that it was John Doe committing the crime, so Doe didn’t know that Doe was doing something wrong. That may be a hard sell to a judge or jury, especially as the Commonwealth will find an expert who will say “Hogwash.”

Similarly, an expert might be willing to say that John Does was acting under an irresistible impulse to commit bank robberies because “Willie Sutton” was in effect compelling him to do so.  Again, this will be a hard sell.

But no expert would have any difficulty in opining that John Doe wouldn’t commit bank robberies if he were not mentally ill — provided they were certain it was true and his belief that he was Sutton not merely an act. Therein lies the difficulty in convincing a court, especially an appellate court, to adopt a “but for” standard — it make mental health evidence too tempting for the defendant to suddenly realize that he must have been crazy.

I think a more likely outcome will be that the Court takes the legislature at its word and decides that the evidence must show that because of the mental illness “the defendant did not have the intent required for the offense charged.”  In other words, the evidence will most likely be used to reduce, rather than eliminate, culpability for an offense.  This is sometimes called “diminished responsibility.” This is most often applied to reduce a charge of murder to manslaughter as, for example, where the defendant believed that the victim was someone who had committed a horrible crime and “deserved” to die, even while knowing that vigilante justice was wrong.  If it was simply a case of mistaken identity, the crime would still be murder; but if the defendant’s mental illness caused him to believe incorrectly that his neighbor was secretly a pedophile or rapist, it might warrant a reduction in the charge and possible a suspended sentence with a civil commitment for treatment.

To reach this point, the law will probably be used by a reform-minded prosecutor to strike a plea bargain with a defendant, and may never reach an appellate court for imprimatur if the trial court is willing to accept the plea and, perhaps an agreed sentence that includes mental health treatment.  I think a trial court will be far less likely to permit a defendant to make such an argument to a jury, and it might reach the appellate courts that way.  Time will tell.